Possible equipment failure - Focus issues.

[deathrox]

I need others opinions on the quality of focus in these images. Camera mounted on a fully retracted tripod sitting on a table sitting on a concrete slab. Shot with 10second self-timer.



I’ll link to the full 6000x4000 images RAW NEF no PP.

To my eyes, which I don’t trust any more, these all appear soft.

http://www.henrywho.xyz/temp/DSC_0409_001.jpg

http://www.henrywho.xyz/temp/DSC_0410_002.jpg

http://www.henrywho.xyz/temp/DSC_0411_003.jpg
(http://www.henrywho.xyz/temp/DSC_0411.NEF)

[Tannin]

My first impression is that the focus is indeed out, and that this isn't really too much of a surprise as I can't see anything in the pictured scene that would give the auto-focus system anything to "grip" on. There are many "half-focusable" objects in view: the wire, the not-quite-flat, shiny plastic, the white wires (too thin for much "grip"), the grain of the wooden post (rounded and smoothed by many coats of paint), and the greenery in the distance (part-obscured by wire). The backlighting just serves to make things more difficult, as does the fact that the objects we want to focus on are dim while the background is bright.

Most (all?) focus systems would find this scene difficult. At a guess, I reckon my 5D II wouldn't have a hope; the 50D might or might not manage, the 7D would probably be OK, and the 1D IV would shrug and say "This is easy, where's the problem?". But that's only a guess, and in any case I don't know what sort of AF system your Nikon has. The key takeaway here is that many a fine camera would struggle with this scene, so don't worry too much about your equipment. Auto-focus is magic but it isn't magic!

I suggest using a single focus point placed over whichever object seems to provide the best "grip" (probably the left-hand edge of the orange post). Failing that, it's back to the old Mark 1 Eyeball.

- - - Updated - - -

PS: from the EXIF I see that you have a 1/6th exposure. I presume that you are using mirror lock-up.

Also, we are at maximum zoom and wide open, both things that stretch the poor little lens to the limit of its ability and should normally be avoided. But this setting is doubtless deliberate as you are worried about the hardware and want to test it.

[deathrox]

This is why I supplied 3 images. The last was focused on the eyelet which should provide excelent oportunity for accurate focus.
No mirror lockup on the D3300.

I just took the gear outside, very little breeze blowing right now. I actually saw the camera move when an image was taken! I knew the tripod was crap but only now realise just how crap!

Not sure what to do now.

After further testing I convinced my 18-55 lens is ok and the camera also. If there is an issue it's with the 55-300 lens only.

[Tannin]

OK. Let's make sure of that before we rush into conclusions. With Canon cameras, if you open the raw file using the officially supplied raw converter (DPP) there is a option to display (a) the focus points, and (b) the focus points which were actually used by the camera to control the AF. Is there a similar facility with the Nikon software? It would be quite useful here as it would show us exactly what the camera is trying to do. If you don't have that facility, the next best thing might be to re-shoot using known, manually-selected focus points.

(I might mention here my experience that a single, manually selected focus point is almost always more likely to achieve correct focus in awkward conditions than several auto-selected ones. Cameras seem to get confused when they have several options to choose from. Advanced (expensive) AF systems do it better than older and cheaper ones, but even with a pro-grade camera, if in doubt, I'll select the AF point myself. Other users, and users of other brands, may or may not have found the same thing.)

Given that we are trying to confirm or eliminate AF issues, let's take the confounding variables out of the equation by using a shutter speed that will clearly eliminate camera shake. (This doesn't look like camera shake to me, but let's make sure.)

Are you having (suspected) focus difficulties more generally? If so, are they always under similar circumstances? (E.g., only with a particular lens, or a particular type of subject, or at low shutter speeds?)

Step by step, we will get to the bottom of this.

- - - Updated - - -

Responding to your edit:


I just took the gear outside, very little breeze blowing right now. I actually saw the camera move when an image was taken! I knew the tripod was crap but only now realise just how crap!


This doesn't look like camera shake to me, but let's make sure.

Hmmmph. Shows how much I know. :(

Pending a better tripod, why not test-shoot at (say) 1/500th or 1/1000th? At 1/1000th, camera movement is irrelevant. Doesn't matter if you need to go to some crazy-high ISO, it's only for testing.

[deathrox]

I shot this with the 18-55.
http://www.henrywho.xyz/temp/DSC_0413_001.jpg

Of course I've been using single point focussing.

Nikon software has "Show focus points" but it does nothing, I'm assuming the D3300 does not dump this info to the EXIF data.


Ha! Just loaded an image taken with the 55-300 and it's showing the focus point.

[John King]

I just took the gear outside, very little breeze blowing right now. I actually saw the camera move when an image was taken! I knew the tripod was crap but only now realise just how crap!

Not sure what to do now.

After further testing I convinced my 18-55 lens is ok and the camera also. If there is an issue it's with the 55-300 lens only.

Get a new tripod! I have several that are as stable as jelly ... I use them for off camera flash.

I also bought a s/h Manfrotto 190 with a Manfrotto tilt/shift head for a good price after discovering how crappy mine were (much as you have done). It is steady enough for my cameras and uses.

After doing that, re-test your lenses ... :nod:.

[deathrox]

Get a new tripod! I have several that are as stable as jelly ... I use them for off camera flash.



Would love to but the nearest camera shop is 3 hours one way. I go once every 3 months.

Check this out.
http://www.henrywho.xyz/temp/Focus%20point.jpg


http://www.henrywho.xyz/temp/DSC_0135_001.jpg

Now 100%

http://www.henrywho.xyz/temp/Focus%20point2.jpg

[John King]

Yeah. Moura to Rocky is a fair way. We used to own a 12,000 acre property near Nipan. Name of Kianga. It's a coal mine now :(.

The good bit is that things like tripods are safe to buy over the Internet. Get some advice here for what would be suitable for your gear and uses.

I just priced a UHS-II memory card from BHPhoto in NY. Delivered it is $90 Aus vs $200 at my local Ted's! Identical card.

- - - Updated - - -

PS: As for your lens question, I'm not familiar with it. However, it doesn't look terrific to me!

[deathrox]

I buy on-line all the time. But tripods incurr large freight charges. My budget is tiny so on top of the issue of needing a new tripod, I simply can't afford, even a half decent, one.

If my lens is having issues I'm up sh*t creek without a paddle.

[John King]

Something like this?

http://www.ebay.com.au/itm/Manfrotto-Tripod-Camera-As-New-Compact-Advanced-Free-Postage/112445980747?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIM.MBE%26ao%3D2%26asc%3D20140106155344%26meid%3D311d41cfd1d449f3828cf745165028df%26pid%3D100005%26rk%3D2%26rkt%3D3%26sd%3D11244 5987285

I don't know if this is suitable for your use??

I suspect that mine is a heavier duty one than this, but at $130 delivered ... Worth looking looking at?

- - - Updated - - -

Or maybe this one?

http://www.ebay.com.au/itm/Manfrotto-Tripod-190PRO-Made-In-Italy-With-3-Way-Head-804RC2-/112445987285?hash=item1a2e4d99d5:g:5rgAAOSwHLNZRKZ6

[ameerat42]

DR. Do you have micro-AF adjustment on that camera?

With your micrometer shot I thought you were doing a micro-AF test, but I see it's
with the non-problematic lens. Do you know how to check if your lens has a back-
or front-focus issue?

- - - Updated - - -

The issue is that the lens focus point is set onto part of the subject but the lens locks focus
a bit before or after the subject. That's what I'm asking. On many cameras you can fix this for
each lens you use. It happened to me with a 30mm prime once. Now it is pin/tack sharp.

[deathrox]

The issue is that the lens focus point is set onto part of the subject but the lens locks focus
a bit before or after the subject.

I've heard of it but even so it should still mean that part of ther image is in perfect focus, I'm not sure if any of my image is in perfect focus.
Either way I have no idea how to adjust it with my D3300.

- - - Updated - - -


With your micrometer shot I thought you were doing a micro-AF test, but I see it's
with the non-problematic lens.

The last few are with the "problematic kens" 55-300. Even with glasses I do not trrust my eyes to judge accurate focus.

[ameerat42]

Yes. For that problem, part of the image should be in focus, just not the part you may want to be.

But may I suggest a test with higher shutter speeds, so eliminating any possible camera shake, though
none hits me particularly in the eye here. At the same elevated shutter speed, change the aperture then
between shots.

For "critical" displaying here, present only a few 100% crops of the subject in question, such as the
"eyelet" above. 100% crop means a crop of the original sized image, withOUT re-sampling it at all.
A "useful" way to get a 100% crop is to display the image at full size on your screen, then judiciously
cropping a part of it. The critical thing, though, is NOT to re-sample/size it.

Back to possible causes for "soft" images, another such would be misaligned optics:eek:

[William W]

I didn’t extensively interrogate all the Images, but the first image was pulled at 1/20th second and the lens was wide open. Tannin (Tony) has already mentioned the lens being wide open and the slow shutter speed, but, a particular point to note is the likelihood of a “beat” vibration being established that will cause camera shake blur at a particular range of slower shutter speeds when the lens is mounted on a tripod.

This blur is a result of movement caused by the mirror slap. I scan-read the whole thread and I think that you mentioned your camera does not have “Mirror Up”. This mirror slap blur (typically) occurs at shutter speeds between ¼ second and 1/30th second, but is also dependent upon the physicality of the lens, tripod, head and mounting system. I have done extensive tests with my Canon Gear and for the heavier lenses (like 70 to 200, 300 and 400) the “worst spot” is 1/15th second. I have not tested the Canon (EF-S) 55 to 250 or 18 to 55 as extensively, however I can report recording blur at the slower end of those shutter speeds, typically around 1/8th second, probably a bit slower for the 18 to 55.

Note that these tests are using well planted, heavy duty tripod with a 3 way Head assembly and no QRP.

BTW I have found two ways of addressing what I term the "mirror slap blur" and one remedy for it, those are:

> Sandbag on top of lens - really very good
> Mirror Up Technique - OK to good - I sometimes can achieve an almost unnoticeable blur, with a long lens at around 1/8th - my guess is shutter curtain movement is creating the vibration

> Don't use anywhere near that range of shutter speeds

*

When testing, another element to consider is “filter” – if you have any filter on the lens remove it for testing.

*

A combination of a ruler at 45 degrees and various day to day objects under different lighting conditions is a good method for testing AF (and recording for each test where exactly the AF was expected to be locked), however, within the above description photographers typically make these tests at shorter rather than longer Subject Distances - it is also a good idea to make a similar test at longer Subject distances and a good long fence makes for a very good outdoor test target - a small party hat or some similar object to sit on one of the fence posts makes for a good target.

*

I think that if you are specifically testing for AF accuracy, then it is NOT a good idea to ONLY test at the lens’s maximum aperture – the reason being, unless one is shooting a test chart, it can be difficult to differentiate between ‘focus inaccuracy’ and ‘soft image due to limit of lens’s optics’: even with shooting the appropriate AF test charts, to an untrained eye the difference can be difficult to pick, especially if the test images are all at one Aperture and/or at one Focal Length.

In the absence of a test chart, a newspaper mounted on a flat wall would be useful to FIRSTLY test the OPTICS of the TWO lenses at different APERTURES and at different FOCAL LENGTHS, before you beging testing the AF accuracy: this approach will at the least ensure an anecdotal appreciation of the overall clarity of the images that you should encounter when testing the AF so that you are less likely to confuse general image quality with an inaccurate AF function.

I encourage the approach to FIRSTLY become familiar with the overall optical performance at different apertures and focal lengths before you test the AF functionality and before you spend money on new tripods or lenses.

WW